virt 3.0 now allows up to 512 vcpus whereas for earlier machine
types, max_cpus was set to 255 and any attempt to start the
machine with vcpus > 255 was rejected at a very early stage,
in vl.c/main level.
512 is the max supported by KVM. Anyway the actual vcpu count
that can be achieved depends on other parameters such as the
acceleration mode, the vgic version, the host kernel version.
Those are discovered later on.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 1529072910-16156-12-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
With this patch, virt-3.0 machine uses a new 256MB ECAM region
by default instead of the legacy 16MB one, if highmem is set
(LPAE supported by the guest) and (!firmware_loaded || aarch64).
Indeed aarch32 mode FW may not support this high ECAM region.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 1529072910-16156-11-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This patch defines a new ECAM region located after the 256GB limit.
The virt machine state is augmented with a new highmem_ecam field
which guards the usage of this new ECAM region instead of the legacy
16MB one. With the highmem ECAM region, up to 256 PCIe buses can be
used.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 1529072910-16156-9-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
With a VGICv3 KVM device, if the number of vcpus exceeds the
capacity of the legacy redistributor region (123 redistributors),
we now attempt to register a second redistributor region. Up to
512 redistributors can fit in this latter on top of the 123 allowed
by the legacy redistributor region.
Registering this second redistributor region is possible if the
host kernel supports the following VGICv3 KVM device group/attribute:
KVM_DEV_ARM_VGIC_GRP_ADDR/KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION.
In case the host kernel does not support the registration of several
redistributor regions and the requested number of vcpus exceeds the
capacity of the legacy redistributor region, the GICv3 device
initialization fails with a proper error message and qemu exits.
At the moment the max number of vcpus still is capped by the
virt machine class max_cpus.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 1529072910-16156-8-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This patch allows the creation of a GICv3 node with 1 or 2
redistributor regions depending on the number of smu_cpus.
The second redistributor region is located just after the
existing RAM region, at 256GB and contains up to up to 512 vcpus.
Please refer to kernel documentation for further node details:
Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.txt
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 1529072910-16156-6-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
To prepare for multiple redistributor regions, we introduce
an array of uint32_t properties that stores the redistributor
count of each redistributor region.
Non accelerated VGICv3 only supports a single redistributor region.
The capacity of all redist regions is checked against the number of
vcpus.
Machvirt is updated to set those properties, ie. a single
redistributor region with count set to the number of vcpus
capped by 123.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 1529072910-16156-4-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
QEMU 3.0 enables strict check for compression & decompression to
make the migration more robust, that depends on the source to fix
the internal design which triggers the unexpected error conditions
To make it work for migrating old version QEMU to 2.13 QEMU, we
introduce this parameter to disable the error check on the
destination which is the default behavior of the machine type
which is older than 2.13, alternately, the strict check can be
enabled explicitly as followings:
-M pc-q35-2.11 -global migration.decompress-error-check=true
Signed-off-by: Xiao Guangrong <xiaoguangrong@tencent.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Remove those unneeded includes to speed up the compilation
process a little bit. (Continue 7eceff5b5a cleanup)
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20180528232719.4721-13-f4bug@amsat.org>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
load_dtb() depends on arm_load_kernel() to figure out place
in RAM where it should be loaded, but it's not required for
arm_load_kernel() to work. Sometimes it's neccesary for
devices added with -device/device_add to be enumerated in
DTB as well, which's lead to [1] and surrounding commits to
add 2 more machine_done notifiers with non obvious ordering
to make dynamic sysbus devices initialization happen in
the right order.
However instead of moving whole arm_load_kernel() in to
machine_done, it's sufficient to move only load_dtb() into
virt_machine_done() notifier and remove ArmLoadKernelNotifier/
/PlatformBusFDTNotifierParams notifiers, which saves us ~90LOC
and simplifies code flow quite a bit.
Later would allow to consolidate DTB generation within one
function for 'mach-virt' board and make it reentrant so it
could generate updated DTB in device hotplug secenarios.
While at it rename load_dtb() to arm_load_dtb() since it's
public now.
Add additional field skip_dtb_autoload to struct arm_boot_info
to allow manual DTB load later in mach-virt and to avoid touching
all other boards to explicitly call arm_load_dtb().
1) (ac9d32e hw/arm/boot: arm_load_kernel implemented as a machine init done notifier)
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 1525691524-32265-4-git-send-email-imammedo@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
platform-bus were using machine_done notifier to get and map
(assign irq/mmio resources) dynamically added sysbus devices
after all '-device' options had been processed.
That however creates non obvious dependencies on ordering of
machine_done notifiers and requires carefull line juggling
to keep it working. For example see comment above
create_platform_bus() and 'straitforward' arm_load_kernel()
had to converted to machine_done notifier and that lead to
yet another machine_done notifier to keep it working
arm_register_platform_bus_fdt_creator().
Instead of hiding resource assignment in platform-bus-device
to magically initialize sysbus devices, use device plug
callback and assign resources explicitly at board level
at the moment each -device option is being processed.
That adds a bunch of machine declaration boiler plate to
e500plat board, similar to ARM/x86 but gets rid of hidden
machine_done notifier and would allow to remove the dependent
notifiers in ARM code simplifying it and making code flow
easier to follow.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Message-id: 1525691524-32265-3-git-send-email-imammedo@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
ARM virt machine now exposes a new "iommu" option.
The SMMUv3 IOMMU is instantiated using -machine virt,iommu=smmuv3.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Prem Mallappa <prem.mallappa@broadcom.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1524665762-31355-15-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Add code to instantiate an smmuv3 in virt machine. A new iommu
integer member is introduced in VirtMachineState to store the type
of the iommu in use.
Signed-off-by: Prem Mallappa <prem.mallappa@broadcom.com>
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1524665762-31355-13-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This allows to pin the host controller in the Linux PCI domain space.
Linux requires that property to be available consistently or not at all,
in which case the domain number becomes unstable on additions/removals.
Adding it here won't make a difference in practice for most setups as we
only expose one controller.
However, enabling Jailhouse on top may introduce another controller, and
that one would like to have stable address as well. So the property is
needed for the first controller as well.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Message-id: 3301c5bc-7b47-1b0e-8ce4-30435057a276@web.de
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Change all the uses of serial_hds[] to go via the new
serial_hd() function. Code change produced with:
find hw -name '*.[ch]' | xargs sed -i -e 's/serial_hds\[\([^]]*\)\]/serial_hd(\1)/g'
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-id: 20180420145249.32435-8-peter.maydell@linaro.org
Instead of using "1.0" as the system version of SMBIOS, we should use
mc->name for mach-virt machine type to be consistent other architectures.
With this patch, "dmidecode -t 1" (e.g., "-M virt-2.12,accel=kvm") will
show:
Handle 0x0100, DMI type 1, 27 bytes
System Information
Manufacturer: QEMU
Product Name: KVM Virtual Machine
Version: virt-2.12
Serial Number: Not Specified
...
instead of:
Handle 0x0100, DMI type 1, 27 bytes
System Information
Manufacturer: QEMU
Product Name: KVM Virtual Machine
Version: 1.0
Serial Number: Not Specified
...
For backward compatibility, we allow older machine types to keep "1.0"
as the default system version.
Signed-off-by: Wei Huang <wei@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 20180322212318.7182-1-wei@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Add support for passing 'max' to -machine gic-version. By analogy
with the -cpu max option, this picks the "best available" GIC version
whether you're using KVM or TCG, so it behaves like 'host' when
using KVM, and gives you GICv3 when using TCG.
Also like '-cpu host', using -machine gic-version=max' means there
is no guarantee of migration compatibility between QEMU versions;
in future 'max' might mean '4'.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20180308130626.12393-7-peter.maydell@linaro.org
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Allow the virt board to support '-cpu max' in the same way
it already handles '-cpu host'.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20180308130626.12393-6-peter.maydell@linaro.org
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Move virt's PSCI DT fixup code to arm/boot.c and set this fixup to
happen automatically for every board that doesn't mark "psci-conduit"
as disabled. This way emulated boards other than "virt" that rely on
PSIC for SMP could benefit from that code.
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Philippe Mathieu-Daudé <f4bug@amsat.org>
Cc: Marcel Apfelbaum <marcel.apfelbaum@zoho.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: qemu-devel@nongnu.org
Cc: qemu-arm@nongnu.org
Cc: yurovsky@gmail.com
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
We were passing a NULL error pointer to the object_property_set_bool()
call that realizes the CPU object. This meant that we wouldn't detect
failure, and would plough blindly on to crash later trying to use a
NULL CPU object pointer. Detect errors and fail instead.
In particular, this will be necessary to detect the user error
of using "-cpu host" without "-enable-kvm" once we make the host
CPU type be registered unconditionally rather than only in
kvm_arch_init().
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Remove dependency of possible_cpus on 1st CPU instance,
which decouples configuration data from CPU instances that
are created using that data.
Also later it would be used for enabling early cpu to numa node
configuration at runtime qmp_query_hotpluggable_cpus() should
provide a list of available cpu slots at early stage,
before machine_init() is called and the 1st cpu is created,
so that mgmt might be able to call it and use output to set
numa mapping.
Use MachineClass::possible_cpu_arch_ids() callback to set
cpu type info, along with the rest of possible cpu properties,
to let machine define which cpu type* will be used.
* for SPAPR it will be a spapr core type and for ARM/s390x/x86
a respective descendant of CPUClass.
Move parse_numa_opts() in vl.c after cpu_model is parsed into
cpu_type so that possible_cpu_arch_ids() would know which
cpu_type to use during layout initialization.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Message-Id: <1515597770-268979-1-git-send-email-imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Replace the TYPE_SYS_BUS_DEVICE entry in the allowed sysbus
device list with the two device types that are really supported
by the virt machine: vfio-amd-xgbe and vfio-calxeda-xgmac.
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: qemu-arm@nongnu.org
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20171125151610.20547-3-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The existing has_dynamic_sysbus flag makes the machine accept
every user-creatable sysbus device type on the command-line.
Replace it with a list of allowed device types, so machines can
easily accept some sysbus devices while rejecting others.
To keep exactly the same behavior as before, the existing
has_dynamic_sysbus=true assignments are replaced with a
TYPE_SYS_BUS_DEVICE entry on the allowed list. Other patches
will replace the TYPE_SYS_BUS_DEVICE entries with more specific
lists of devices.
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alexander Graf <agraf@suse.de>
Cc: David Gibson <david@gibson.dropbear.id.au>
Cc: Stefano Stabellini <sstabellini@kernel.org>
Cc: Anthony Perard <anthony.perard@citrix.com>
Cc: qemu-arm@nongnu.org
Cc: qemu-ppc@nongnu.org
Cc: xen-devel@lists.xenproject.org
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20171125151610.20547-2-ehabkost@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Calculating default node-ids for CPUs in possible_cpu_arch_ids()
is rather fragile since defaults calculation uses nb_numa_nodes but
callback might be potentially called early before all -numa CLI
options are parsed, which would lead to cpus assigned only upto
nb_numa_nodes at the time possible_cpu_arch_ids() is called.
Issue was introduced by
(7c88e65 numa: mirror cpu to node mapping in MachineState::possible_cpus)
and for example CLI:
-smp 4 -numa node,cpus=0 -numa node
would set props.node-id in possible_cpus array for every non
explicitly mapped CPU to the first node.
Issue is not visible to guest nor to mgmt interface due to
1) implictly mapped cpus are forced to the first node in
case of partial mapping
2) in case of default mapping possible_cpu_arch_ids() is
called after all -numa options are parsed (resulting
in correct mapping).
However it's fragile to rely on late execution of
possible_cpu_arch_ids(), therefore add machine specific
callback that returns node-id for CPU and use it to calculate/
set defaults at machine_numa_finish_init() time when all -numa
options are parsed.
Reported-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1496314408-163972-1-git-send-email-imammedo@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
there are 2 use cases to deal with:
1: fixed CPU models per board/soc
2: boards with user configurable cpu_model and fallback to
default cpu_model if user hasn't specified one explicitly
For the 1st
drop intermediate cpu_model parsing and use const cpu type
directly, which replaces:
typename = object_class_get_name(
cpu_class_by_name(TYPE_ARM_CPU, cpu_model))
object_new(typename)
with
object_new(FOO_CPU_TYPE_NAME)
or
cpu_generic_init(BASE_CPU_TYPE, "my cpu model")
with
cpu_create(FOO_CPU_TYPE_NAME)
as result 1st use case doesn't have to invoke not necessary
translation and not needed code is removed.
For the 2nd
1: set default cpu type with MachineClass::default_cpu_type and
2: use generic cpu_model parsing that done before machine_init()
is run and:
2.1: drop custom cpu_model parsing where pattern is:
typename = object_class_get_name(
cpu_class_by_name(TYPE_ARM_CPU, cpu_model))
[parse_features(typename, cpu_model, &err) ]
2.2: or replace cpu_generic_init() which does what
2.1 does + create_cpu(typename) with just
create_cpu(machine->cpu_type)
as result cpu_name -> cpu_type translation is done using
generic machine code one including parsing optional features
if supported/present (removes a bunch of duplicated cpu_model
parsing code) and default cpu type is defined in an uniform way
within machine_class_init callbacks instead of adhoc places
in boadr's machine_init code.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <1505318697-77161-6-git-send-email-imammedo@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@xilinx.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Let's provide the GPEX host bridge with the INTx/gsi mapping. This is
needed for INTx/gsi routing.
Signed-off-by: Pranavkumar Sawargaonkar <pranavkumar@linaro.org>
Signed-off-by: Tushar Jagad <tushar.jagad@linaro.org>
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Tested-by: Feng Kan <fkan@apm.com>
Message-id: 1505296004-6798-3-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
If a KVM PMU init or set-irq attr call fails we just silently stop
the PMU DT node generation. The only way they could fail, though,
is if the attr's respective KVM has-attr call fails. But that should
never happen if KVM advertises the PMU capability, because both
attrs have been available since the capability was introduced. Let's
just abort if this should-never-happen stuff does happen, because,
if it does, then something is obviously horribly wrong.
Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Message-id: 1500471597-2517-5-git-send-email-drjones@redhat.com
[PMM: change kvm32.c kvm_arm_pmu_init() to the new API too]
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Move the in-kernel-irqchip test to only guard the set-irq
stage, not the init stage of the PMU. Also add the PMU to
the KVM device irq line synchronization to enable its use.
Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Message-id: 1500471597-2517-4-git-send-email-drjones@redhat.com
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
When adding a PMU with a userspace irqchip we skip the set-irq
stage of device creation. Split the 'create' function into two
functions 'init' and 'set-irq' so they may be called separately.
Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Message-id: 1500471597-2517-3-git-send-email-drjones@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Mimicking gicv3-maintenance-interrupt, add the PMU's interrupt to
CPU state.
Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1500471597-2517-2-git-send-email-drjones@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Add virt-2.10 machine type.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 1502106581-11714-1-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Use the new functions memory_region_init_{ram,rom,rom_device}()
instead of manually calling the _nomigrate() version and then
vmstate_register_ram_global().
Patch automatically created using coccinelle script:
spatch --in-place -sp_file scripts/coccinelle/memory-region-init-ram.cocci -dir hw
(As it turns out, there are no instances of the rom and
rom_device functions that are caught by this script.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 1499438577-7674-8-git-send-email-peter.maydell@linaro.org
Rename memory_region_init_ram() to memory_region_init_ram_nomigrate().
This leaves the way clear for us to provide a memory_region_init_ram()
which does handle migration.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 1499438577-7674-4-git-send-email-peter.maydell@linaro.org
This is based on patch Shannon Zhao originally posted.
Cc: Shannon Zhao <zhaoshenglong@huawei.com>
Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Shannon Zhao <shannon.zhao@linaro.org>
Message-id: 20170529173751.3443-3-drjones@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-Id: <1494415802-227633-11-git-send-email-imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
it will allow switching from cpu_index to property based
numa mapping in follow up patches.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-Id: <1494415802-227633-5-git-send-email-imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Originally CPU threads were by default assigned in
round-robin fashion. However it was causing issues in
guest since CPU threads from the same socket/core could
be placed on different NUMA nodes.
Commit fb43b73b (pc: fix default VCPU to NUMA node mapping)
fixed it by grouping threads within a socket on the same node
introducing cpu_index_to_socket_id() callback and commit
20bb648d (spapr: Fix default NUMA node allocation for threads)
reused callback to fix similar issues for SPAPR machine
even though socket doesn't make much sense there.
As result QEMU ended up having 3 default distribution rules
used by 3 targets /virt-arm, spapr, pc/.
In effort of moving NUMA mapping for CPUs into possible_cpus,
generalize default mapping in numa.c by making boards decide
on default mapping and let them explicitly tell generic
numa code to which node a CPU thread belongs to by replacing
cpu_index_to_socket_id() with @cpu_index_to_instance_props()
which provides default node_id assigned by board to specified
cpu_index.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <1494415802-227633-2-git-send-email-imammedo@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Currently cpu_index is implicitly auto assigned during
cpu.realize() time cpu_exec_realizefn()->cpu_list_add().
It happens to match index in possible_cpus so take
control over it and make board initialize cpu_index
to possible_cpus index explicitly. It will at least
document that board is in control of it and when
'-device cpu' support comes it will keep cpu_index
stable regardless of order cpus are created so it won't
break migration.
Within this series it will be used for internal
conversion from storing cpu_index based NUMA node
bitmaps to property based mapping with possible_cpus,
And will allow map cpu_index to a CPU entry in
possible_cpus array.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-Id: <1493816238-33120-5-git-send-email-imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
for now precalculate and store mp_afinity in possible_cpus
as ARM cpus don't have socket/core/thread-id properties yet.
In follow patches possible_cpus will be used for storing
and setting NUMA node mapping and replace legacy bitmap
based numa_info[node_id].node_cpu/numa_get_node_for_cpu()
For the lack of better idea, this patch cannibalizes
possible_cpus.cpus[x].props.thread_id so that
*_cpu_index_to_props() callback could return addressable
by props CPU which will be used by machine_set_cpu_numa_node()
in follow up patches to assign a CPU to node. But
cannibalizing is fine for now as that thread_id isn't exposed
to users (no hotpluggable_cpus callback support for ARM yet)
and it will be used only internally until 'device_add cpu'
is supported where we can decide on which properties to use.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1493816238-33120-4-git-send-email-imammedo@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1493816238-33120-3-git-send-email-imammedo@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
In 2.9 ITS will block save/restore and migration use cases. As such,
let's introduce a user option that allows to turn its instantiation
off, along with GICv3. With the "its" option turned false, migration
will be possible, obviously at the expense of MSI support (with GICv3).
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Message-id: 1487681108-14452-1-git-send-email-eric.auger@redhat.com
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
object_new(FOO) returns an object with ref_cnt == 1
and following
object_property_set_bool(cpuobj, true, "realized", NULL)
set parent of cpuobj to '/machine/unattached' which makes
ref_cnt == 2.
Since machvirt_init() doesn't take ownership of cpuobj
returned by object_new() it should explicitly drop
reference to cpuobj when dangling pointer is about to
go out of scope like it's done pc_new_cpu() to avoid
object leak.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-id: 1487253461-269218-1-git-send-email-imammedo@redhat.com
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Fw-cfg recently learned how to directly access guest memory and does so in
cache coherent fashion. Tell the guest about that fact when it's using DT.
Signed-off-by: Alexander Graf <agraf@suse.de>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: Shannon Zhao <shannon.zhao@linaro.org>
Message-id: 1486644810-33181-5-git-send-email-agraf@suse.de
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
QEMU emulated hardware is always dma coherent with its guest. We do
annotate that correctly on the PCI host controller, but left out
virtio-mmio.
Recent kernels have started to interpret that flag rather than take
dma coherency as granted with virtio-mmio. While that is considered
a kernel bug, as it breaks previously working systems, it showed that
our dt description is incomplete.
This patch adds the respective marker that allows guest OSs to evaluate
that our virtio-mmio devices are indeed cache coherent.
Signed-off-by: Alexander Graf <agraf@suse.de>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Message-id: 1486644810-33181-2-git-send-email-agraf@suse.de
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
